Abstract: This project focuses on vehicle detection and counting for traffic control, a crucial task for highway regulation. Detecting and counting vehicles accurately is challenging due to their varying structures. The paper presents a video-based approach using OpenCV technologies, specifically background subtraction to identify moving vehicles. It also applies techniques like thresholding, adaptive morphology operations, and hole filling to improve detection accuracy. Vehicle counting is performed using virtual identification zones. The experimental results show that the proposed system achieves an accuracy of about 96%.

| DOI: 10.17148/IJIREEICE.2025.13426